
Chapter 4. Floating-Point Unit (FPU)
For More Information On This Product,
Go to: www.freescale.com
4-9
FPU Programmer’s Model
4.3.2 Floating-Point Status Register (FPSR)
The FPSR, Figure 4-10, contains a floating-point condition code byte (FPCC), a
floating-point exception status byte (EXC), and a floating-point accrued exception byte
(AEXC). The user can read or write all FPSR bits. Execution of most floating-point
instructions modifies FPSR. FPSR is loaded using FMOVE or FRESTORE. A processor
reset or a restore operation of the null state clears the FPSR.
Figure 4-10. Floating-Point Status Register (FPSR)
Table 4-5 describes FPSR fields.
7–0
MC
Mode control byte. Control FPU operating modes.
7
—
Reserved, should be cleared.
6
PREC
Rounding precision
0 Double (D)
1 Single (S)
5–4
RND
Rounding mode
00 To nearest (RN)
01 To zero (RZ)
10 To minus infinity (RM)
11 To plus infinity (RP)
3–0
—
Reserved, should be cleared.
31 28 27 26 25
24
23
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2 0
Field
—
N Z
I
NAN
—
BSUN INAN OPERR OVFL UNFL DZ INEX IDE IOP OVFL UNFL DZ INEX —
Reset
All zeros
R/W
R/W
Table 4-5. FPSR Field Descriptions
Bits
Field
Description
31–24
FPCC
Floating-point condition code byte. Contains four condition code bits that are set after completion
of all arithmetic instructions involving the floating-point data registers. The floating-point store
operation, FMOVEM, and move system control register instructions do not affect the FPCC.
31–28
Reserved, should be cleared.
27
N
Negative
26
Z
Zero
25
I
Infinity
24
NAN
Not-a-number
23–16
—
Reserved, should be cleared.
Table 4-4. FPCR Field Descriptions (Continued)
Bits
Field
Description
FPCC
Exception Status Byte (EXC)
AEXC Byte
F
Freescale Semiconductor, Inc.
n
.